From: Yingya Han <yingyax.han@intel.com>
To: dts@dpdk.org
Cc: Yingya Han <yingyax.han@intel.com>
Subject: [dts][PATCH V2 1/3]test_plans: add ice_iavf_checksum_perf test plan
Date: Fri, 21 Apr 2023 18:29:21 +0000 [thread overview]
Message-ID: <20230421182923.179866-2-yingyax.han@intel.com> (raw)
In-Reply-To: <20230421182923.179866-1-yingyax.han@intel.com>
Signed-off-by: Yingya Han <yingyax.han@intel.com>
---
.../ice_iavf_checksum_perf_test_plan.rst | 224 ++++++++++++++++++
test_plans/index.rst | 1 +
2 files changed, 225 insertions(+)
create mode 100644 test_plans/ice_iavf_checksum_perf_test_plan.rst
diff --git a/test_plans/ice_iavf_checksum_perf_test_plan.rst b/test_plans/ice_iavf_checksum_perf_test_plan.rst
new file mode 100644
index 00000000..891a6536
--- /dev/null
+++ b/test_plans/ice_iavf_checksum_perf_test_plan.rst
@@ -0,0 +1,224 @@
+.. SPDX-License-Identifier: BSD-3-Clause
+ Copyright(c) 2023 Intel Corporation
+
+===============================================================
+Benchmark the performance of iavf checksum forwarding with E810
+===============================================================
+
+Description
+===========
+
+Prerequisites
+=============
+
+1. Hardware:
+
+ 1.1) vf checksum perf test for Intel® Ethernet Network Adapter E810-CQDA2:
+ 1 NIC or 2 NIC cards attached to the same processor and 1 port used of each NIC.
+ 1.2) vf checksum perf test for Intel® Ethernet Network Adapter E810-XXVDA4:
+ 1 NIC card attached to the processor and 4 ports used.
+
+2. Software::
+
+ dpdk: git clone http://dpdk.org/git/dpdk
+ trex: git clone http://trex-tgn.cisco.com/trex/release/v2.93.tar.gz
+
+
+Test Case
+=========
+The test case check the throughput result with ipv4, in the case,
+we will send the bi-direction flows with line rate, then we can check the
+passthrough rate.
+
+Common Steps
+------------
+1. Bind PF ports to kernel driver(ice), then create 1 VF from each PF,
+ take E810-CQDA2 for example::
+
+ echo 1 > /sys/bus/pci/devices/0000\:17\:00.0/sriov_numvfs
+ echo 1 > /sys/bus/pci/devices/0000\:4b\:00.0/sriov_numvfs
+
+2. Ip link set VF trust on and spoofchk off on DUT::
+
+ ip link set $PF_INTF vf 0 trust on
+ ip link set $PF_INTF vf 0 spoofchk off
+
+3. Set vf mac address::
+
+ ip link set ens5f0 vf 0 mac 00:12:34:56:78:01
+ ip link set ens6f0 vf 0 mac 00:12:34:56:78:02
+
+4. Bind tested ports to vfio-pci::
+
+ <dpdk_dir>#./usertools/dpdk-devbind.py -s
+ 0000:17:00.0 'Device 1592' if=ens5f0 drv=ice unused=vfio-pci
+ 0000:4b:00.0 'Device 1592' if=ens6f0 drv=ice unused=vfio-pci
+ <dpdk_dir>#./usertools/dpdk-devbind.py -b vfio-pci <pci device id>
+ <dpdk_dir>#./usertools/dpdk-devbind.py -b vfio-pci 0000:17:01.0
+ <dpdk_dir>#./usertools/dpdk-devbind.py -b vfio-pci 0000:4b:01.0
+
+5. Configure traffic generator to send traffic
+
+ Test flow MAC table.
+
+ +------+---------+------------+----------------+
+ | Flow | Traffic | MAC | MAC |
+ | | Gen. | Src. | Dst. |
+ | | Port | Address | Address |
+ +======+=========+============+================+
+ | 1 | TG0 | Random MAC | 00:12:34:56:01 |
+ +------+---------+------------+----------------+
+ | 2 | TG1 | Random Mac | 00:12:34:56:02 |
+ +------+---------+------------+----------------+
+
+ The Flow IP table.
+
+ +------+---------+------------+------------+
+ | Flow | Traffic | IPV4 | IPV4 |
+ | | Gen. | Src. | Dest. |
+ | | Port | Address | Address |
+ +======+=========+============+============+
+ | 1 | TG0 | Any IP | 198.18.1.0 |
+ +------+---------+------------+------------+
+ | 2 | TG1 | Any IP | 198.18.0.0 |
+ +------+---------+------------+------------+
+
+ Set the packet length : 128 bytes-1518 bytes
+ The IPV4 Dest Address increase with the num 1024.
+
+6. Test results table.
+
+ +-----------+------------+-------------+---------+
+ | Fwd_core | Frame Size | Throughput | Rate |
+ +===========+============+=============+=========+
+ | 1C/1T | 128 | xxxxx Mpps | xxx % |
+ +-----------+------------+-------------+---------+
+ | 1C/1T | ... | xxxxx Mpps | xxx % |
+ +-----------+------------+-------------+---------+
+
+Test Case 1: test_perf_enable_SW_checksum_offload
+-------------------------------------------------
+test steps
+~~~~~~~~~~
+1. Bind VF ports to dpdk driver as common step 4::
+
+ ./usertools/dpdk-devbind.py -b vfio-pci 17:01.0 4b:01.0
+
+2. Start dpdk-testpmd::
+
+ <build_dir>/app/dpdk-testpmd -l 5,6 -n 8 --force-max-simd-bitwidth=512 \
+ -- -i --portmask=0x3 --rxq=1 --txq=1 --txd=1024 --rxd=1024 --forward=io \
+ --nb-cores=1 --enable-rx-cksum
+
+ Note:
+ -force-max-simd-bitwidth: set 512
+ -enable-rx-cksum: enable rx checksum
+
+3. Configure traffic generator to send traffic as common step 5.
+
+4. Record Test results as common step 6.
+
+5. Config SW checksum offload::
+
+ testpmd>stop
+ testpmd>set fwd csum
+ testpmd>port stop all
+ testpmd>csum set ip sw 0
+ testpmd>csum set ip sw 1
+ testpmd>port start all
+ testpmd>start
+
+6. Configure traffic generator to send traffic as common step 5.
+
+ Set the incorrect ipv4 checksum.
+
+7. Record Test results as common step 6.
+
+8. Set the traffic flow as below:
+
+ Ether / IPv4 / UDP / VXLAN / Ether / IPv4 / UDP / payload
+
+ Set the packet length : 128 bytes-1518 bytes.
+
+ Set the inner and outer incorrect ipv4 checksum and udp checksum.
+
+9. Record Test results as common step 6.
+
+
+Test Case 2: test_perf_enable_HW_checksum_offload
+-------------------------------------------------
+test steps
+~~~~~~~~~~
+
+1. Bind VF ports to dpdk driver as common step 4::
+
+ ./usertools/dpdk-devbind.py -b vfio-pci 17:01.0 4b:01.0
+
+2. Start dpdk-testpmd::
+
+ <build_dir>/app/dpdk-testpmd -l 5,6 -n 8 --force-max-simd-bitwidth=512 \
+ -- -i --portmask=0x3 --rxq=1 --txq=1 --txd=1024 --rxd=1024 --forward=io \
+ --nb-cores=1 --enable-rx-cksum
+
+ Note:
+ -force-max-simd-bitwidth: set 512
+ -enable-rx-cksum: enable rx checksum
+
+3. Config HW checksum offload::
+
+ testpmd>stop
+ testpmd>set fwd csum
+ testpmd>port stop all
+ testpmd>csum set ip hw 0
+ testpmd>csum set ip hw 1
+ testpmd>port start all
+ testpmd>set promisc all on
+ testpmd>start
+
+4. Configure traffic generator to send traffic as common step 5.
+
+ Set the incorrect ipv4 checksum.
+
+5. Record Test results as common step 6.
+
+6. Set the traffic flow as below:
+
+ Ether / IPv4 / UDP / VXLAN / Ether / IPv4 / UDP / payload
+
+ Set the packet length : 128 bytes-1518 bytes.
+
+ Set the inner and outer incorrect ipv4 checksum and udp checksum.
+
+7. Record Test results as common step 6.
+
+8. Set HW inner and outer checksum offload::
+
+ testpmd>stop
+ testpmd>port stop all
+ testpmd>csum set udp hw 0
+ testpmd>csum set udp hw 1
+ testpmd>csum set outer-ip hw 0
+ testpmd>csum set outer-ip hw 1
+ testpmd>csum set outer-udp hw 0
+ testpmd>csum set outer-udp hw 1
+ testpmd>csum parse-tunnel on 0
+ testpmd>csum parse-tunnel on 1
+ testpmd>port start all
+ testpmd>set promisc all on
+ testpmd>start
+
+9. Configure traffic generator to send traffic as common step 5.
+
+ Set the incorrect ipv4 checksum.
+
+10. Record Test results as common step 6.
+
+11. Set the traffic flow as below:
+
+ Ether / IPv4 / UDP / VXLAN / Ether / IPv4 / UDP / payload
+
+ Set the packet length : 128 bytes-1518 bytes.
+
+ Set the inner and outer incorrect ipv4 checksum and udp checksum.
+
+12. Record Test results as common step 6.
diff --git a/test_plans/index.rst b/test_plans/index.rst
index 2008f50b..3a91799e 100644
--- a/test_plans/index.rst
+++ b/test_plans/index.rst
@@ -43,6 +43,7 @@ The following are the test plans for the DPDK DTS automated test system.
ice_fdir_test_plan
ice_rx_timestamp_test_plan
ice_ip_fragment_rte_flow_test_plan
+ ice_iavf_checksum_perf_test_plan
ice_iavf_flow_subscribe_test_plan
ice_iavf_ip_fragment_rte_flow_test_plan
ice_iavf_packet_pacing_test_plan
--
2.34.1
next prev parent reply other threads:[~2023-04-21 10:32 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-04-21 18:29 [dts][PATCH V2 0/3] add ice_iavf_checksum_perf test suite Yingya Han
2023-04-21 18:29 ` Yingya Han [this message]
2023-04-21 18:29 ` [dts][PATCH V2 2/3]tests: add ice_iavf_checksum_perf test script Yingya Han
2023-04-21 18:29 ` [dts][PATCH V2 3/3]conf: add ice_iavf_checksum_perf configuration file Yingya Han
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20230421182923.179866-2-yingyax.han@intel.com \
--to=yingyax.han@intel.com \
--cc=dts@dpdk.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).